Generating expanded responses that guide continuance of a human-to computer dialog that is facilitated by a client device and that is between at least one user and an automated assistant. The expanded responses are generated by the automated assistant in response to user interface input provided by the user via the client device, and are caused to be rendered to the user via the client device, as a response, by the automated assistant, to the user interface input of the user. An expanded response is generated based on at least one entity of interest determined based on the user interface input, and is generated to incorporate content related to one or more additional entities that are related to the entity of interest, but that are not explicitly referenced by the user interface input.
Legal claims defining the scope of protection, as filed with the USPTO.
2. The method of claim 1, wherein the trained machine learning model is personalized to the user based on training, of the machine learning model, using training instances that are specific to the user.
3. The method of claim 1, wherein the historical inputs are processed, using the trained machine learning model, in generating the output.
4. The method of claim 3, wherein the historical inputs include a frequency of the user explicitly indicating a desire for a response for further engagement and/or an average duration of past dialogs of the user.
5. The method of claim 4, wherein the dialog inputs are processed, using the trained machine learning model, in generating the output.
6. The method of claim 5, wherein the user interface input is voice input and the dialog inputs include pitch, intonation, and/or tone of the voice input.
7. The method of claim 1, wherein the user interface input identifies an entity of interest, and the prompt for further engagement identifies an additional entity that is associated with the entity of interest.
8. The method of claim 1, wherein the trained generative model is a sequence-to-sequence model.
9. The method of claim 1, wherein the multiple items of responsive content include a first item of responsive content from a first content agent and a second item of responsive content from a second content agent.
10. The method of claim 1, wherein the multiple items of responsive content include a first textual snippet and a second textual snippet.
11. The method of claim 1, wherein the multiple items of responsive content include a textual snippet and an image.
12. The method of claim 1, wherein the multiple items of responsive content include a first item from a news title of the different resources and a second content item from an online forum of the different resources.
13. The method of claim 1, wherein the multiple items of responsive content are selected from a group of candidate items that are returned from one or more disparate content agents.
14. The method of 1, wherein the multiple items of responsive content are applied, on a token-by-token basis, to the trained generative model.
15. The method of 14, wherein the multiple items of responsive content applied to the trained generative model collectively have a first length and wherein the expanded response has a shortened length that is shortened relative to the first length.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
January 28, 2022
January 9, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.